No date

Session Item

Sunday

November 29

08:45 - 10:00

Clinical Stream 2

The role of radiotherapy in locally advanced pancreatic cancer

Overview: The role of radiation therapy in the treatment of pancreatic cancer has been a matter of debate for decades. In this session, Florence Huguet will discuss the optimal target volume based on adequate imaging and pattern of failure analyses. Maria Hawkins will summarize current fractionation regimes including (extreme) hypofractionation. Piero Fossati will give an overview on the specific properties of particle beams and review the currently available evidence in locally-advanced pancreatic cancer. Finally, Thomas Brunner will discuss the optimal integration of radiation therapy into systemic approaches focusing on efficacy and toxicity of sequential and concurrent chemoradiation regimes in different settings.

Session Type: Symposium

Track: Clinical

Journey:

15:55 - 16:00

Looking Beyond D90: Correlating EUBED, gBEUD with Outcome in Cervical Cancer Brachytherapy

Presentation Number: PD-0659

Abstract

Abstract Title:

Looking Beyond D90: Correlating EUBED, gBEUD with Outcome in Cervical Cancer Brachytherapy

(1)Virginia Commonwealth University, Radiation Oncology, Richmond- VA, USA;(2)British Columbia Cancer Center, Radiation Oncology, Kelowna, Canada;(3)London Health Sciences Center, Radiation Oncology, London, Canada;

Show Affiliations

Purpose or Objective

The current formalism for reporting and adding the effects of BT and external beam RT doses is based on a single metric, D90 to CTV_HR and its EQD2. Our previously published work however, has demonstrated that BED(D90) is unable to distinguish between plans that are optimized differently or created with different applicator systems. In comparison, equivalent uniform BED (EUBED) and generalized biologically equivalent uniform dose (gBEUD), which integrates dose inhomogeneity, are better quantities for discerning these important differences. We sought to measure local control in women being treated with radiotherapy for locally advanced cervix cancer and assess its correlation with EUBED and gBEUD integrated over the whole volume of CTV_HR compared to the traditional, single point metrics D90.

Material and Methods

163 women with FIGO 2018 IB3-IIIC2 cervical cancer from 3 institutions were included. All women received EBRT ± chemotherapy and 3-5 fractions of intracavitary brachytherapy. The median follow up for all women was 59.5 months (range 0-87 months). At this time 6.1% have had local recurrences, 12.2% distant recurrences, 1.2% both local and distant. Data was extracted from TPS and dosimetric and radiobiological quantities computed. EUBED and gBEUD were calculated for each fraction after a voxel-based conversion from dose to BED (α/β=10Gy). EUBED uses an extra parameter α (range 0.05-0.5Gy^-1) with smaller values typically associated with radioresistant tumors. Similarly, gBEUD uses a parameter a which allows a variable emphasis on hot/cold spots (range –5,5). Distributions were compared using the 2-sample Kolmogorov-Smirnoff test at 5% significance.

Results

Distributions of all dosimetric quantities were compared among institutions, starting with CTV size and D90:

D90 shows excellent dosimetric quality across all sites and treatments. Two populations were created for each site, patients with no recurrences and patients with local recurrences. The two populations were compared using all variables. No dosimetric variable was able to distinguish between the two populations, with the exception of gBEUD for a positive exponent between 3 and 5, at high levels of statistical significance p- (0.027, 0.038). No such distinction between populations was found when testing for distant recurrences.

Conclusion

D90 did not distinguish between patient populations with different outcomes. EUBED integrates biological effect over a volume in a manner very similar with gBEUD for negative a. Such exponents are typically used in relation with Local Control emphasizing ‘cold spots’. Interestingly, a large ‘hot’ volume is present in all T&O and T&R applications and some practitioners even emphasize contiguity of 150% or 200% isosurfaces in their plans. gBEUD for positive values [3..5] proved to be the only metric that successfully predicted patients with no recurrence. Based on these findings, gBEUD should be included routinely to the set of dosimetric parameters used to evaluate implants.